AI RESEARCH

From Competition to Synergy: Unlocking Reinforcement Learning for Subject-Driven Image Generation

arXiv CS.LG

ArXi:2510.18263v2 Announce Type: replace Subject-driven image generation models face a fundamental trade-off between identity preservation (fidelity) and prompt adherence (editability). While online reinforcement learning (RL), specifically GPRO, offers a promising solution, we find that a naive application of GRPO leads to competitive degradation, as the simple linear aggregation of rewards with static weights causes conflicting gradient signals and a misalignment with the temporal dynamics of the diffusion process.